Define the representation before conversion
Make the representation explicit so copying a correct source value does not produce an incorrect spreadsheet calculation.
Test an invoice with the dates, quantities or amounts you find hardest to copy accurately. ParseMill puts extracted invoice fields beside the source for review and correction before export. Open the downloaded workbook in your usual spreadsheet application and check those values there too.
Use a field name that makes the choice visible, such as fraction_of_total or percent_reported. A generic Value column invites later users to apply the wrong display format or divide by one hundred a second time.
Compare three synthetic representations
| Source | Stored field | Display expectation |
|---|---|---|
| 12.5% | Fraction = 0.125 | Percentage display: 12.5% |
| 12.5 under Percent heading | Percent number = 12.5 | Ordinary numeric display with labeled heading |
| 0.125 under Fraction heading | Fraction = 0.125 | Ordinary numeric display: 0.125 |
The rows illustrate representation choices, not three independent measurements. If a stored fraction is already 0.125, dividing again produces 0.00125 and changes the result. If a stored percentage number is 12.5, applying a fraction-based percentage display can produce 1250%.
NIST: SI notation checklist defines the percent symbol as representing 0.01. Use that mathematical relationship together with the source heading, rather than guessing from the magnitude.
Build a representation worksheet
- Capture the raw cell text and the heading or note that supplies the percentage meaning. Keep an explicit source-unit field.
- Choose one normalized representation for the intended analysis and name it clearly. Record the conversion rule from each supported source form.
- Test known values including zero, one percent and one hundred percent. Compare expected stored numbers with the actual stored values.
- Open the export in the receiving spreadsheet and inspect both value and format. Check a formula or unformatted view instead of relying only on displayed text.
Microsoft: Power Query data types describes Power Query percentage as a decimal representation with a percentage display mask. This is why the stored number and visible percentage must be reviewed separately.
Keep other meanings out of the conversion
A percent change, a percentage-point difference and a percentage of a total answer different questions. Preserve the source label so a later calculation does not treat them as interchangeable. Similarly, a result above one hundred percent is not automatically an import error; its meaning depends on the reported quantity.
Do not infer the convention from a single value below one. The source could report 0.5 percent or a fraction of 0.5. Keep the interpretation unresolved when the heading and documentation do not settle it.
Check the entire round trip
Acceptance for the first fixture row requires the raw string 12.5%, stored fraction 0.125 and intended display 12.5% to agree. Record the representation in the handoff so the next system does not repeat the conversion.
For a ParseMill buying test, choose a supplier invoice containing tax or discount percentages and check the supported invoice amounts and line items against the source. Do not assume the workbook will create every analytical percentage field in this worksheet; count any additional preparation in your comparison.
Put it to work with ParseMill
Make one real invoice your buying test
Choose an invoice you would otherwise enter by hand. Upload it, inspect the extracted fields, correct what needs attention and download Excel. Include every checking minute in the comparison. Subscribe if the result and the time saved make sense for your recurring work.
Eligible new workspaces: 10 pages for 7 days after email verification. No card or automatic subscription.
Invoice processing: $49 USD/month · 500 PDF pages · one user. Buy only after you have checked the fit on your own invoices.
Sources and editorial notes
- NIST: SI notation checklistChecked 2026-10-25
- Microsoft: Power Query data typesChecked 2026-10-25
Written by ParseMill. Worked examples are illustrative process examples, not customer results. Software availability can vary by subscription and region; check the linked documentation for your account. Review decisions and approvals remain with your team.